We have studied the histone deacetylase inhibitors romidepsin and belinostat in both the clinic and in the laboratory. We first pursued the use of depsipeptide/romidepsin as an orphan drug in T cell lymphoma, using both laboratory and clinical strategies. We have extended this work into solid tumors with the hydroxamic acid derivative belinostat. Our multi-institutional clinical trial for cutaneous and peripheral T cell lymphoma (CTCL and PTCL) completed accrual at 131 patients in 6 cohorts. Papers detailing responses to romidepsin in both CTCL and PTCL are published. The responses to depsipeptide are at times dramatic and have been very durable. As an example, one patient received therapy for 2 years, and has remained in complete remission off of therapy for 10 years. Another patient with PTCL remained in continuous complete remission for 5 years before relapse occurred and no other therapy was able to control the disease. The major response rate in cutaneous T cell lymphoma in both our NCI trial and in the Gloucester registration trial was 34-35%. For PTCL, our response rate was 38%. It is important to note that durable responses were also obtained in both subsets of patients by extramural investigators who were participating in our Phase II trial among more than 9 multicenter sites included in the study. These sites included North Shore University Hospital in Manhasset, New York; City of Hope National Cancer Center in Duarte, California; and Peter MacCallum Cancer Center in Melbourne, Australia. Our NCI Phase II trial had a major second objective in addition to evaluating efficacy. That was confirmation of the safety of the agent. EKG abnormalities have been noted following treatment and a great deal of effort has gone into demonstrating the lack of myocardial damage associated with administration of this agent. We analyzed over 4000 ECGs, and collected much ancillary cardiac safety data. The trial also had a significant translational correlative sample component, and we were able to show reproducible increased histone acetylation, and induction of gene expression in normal and tumor cells following romidepsin infusion. Our data suggest that the 24hr timepoint of histone acetylation in peripheal blood mononuclear cells is dually associated with pharmacokinetic parameters including clearance and area under the curve and with disease response. Taken together these data suggest that drug exposure may be important for romidepsin and potentially for the entire class of histone deacetylase inhibitors. Additionally, data have been retrieved from cDNA arrays on samples sent as per protocol to Dr. Louise Showe at University of Pennsylvania. We reported on a Phase I trial of romidepsin on a day 1, 3, and 5 schedule to achieve a more continuous drug effect. We have nearly completed a combination clinical trial with a novel histone deacetylase inhibitor, belinostat, evaluating a 48 hr continuous infusion in combination with cisplatin and etoposide. This trial is based on preclinical evidence of synergy between HDAC inhibitors and chemotherapeutics, when properly scheduled. This was carried out as a Phase I trial in an advanced disease population; we are currently refining a Phase II dose. The Phase II dose will be explored in the same trial design in the small cell lung cancer patient population. We have made a major effort in the last year to understand mechanisms of HDI resistance. This led us to the generation of cell lines with non-Pgp-mediated romidepsin resistance and we have identified increased MEK signaling as a mechanism of resistance. This is apparently mediated via loss of BIM, a proapoptotic mitochondrial protein. We have detailed laboratory studies that show in approach after approach that the mitochondrial apoptotic milieu is critical to responsiveness to romidepsin. We continue to be interested in the striking mechanism of action of the HDIs. At least 5 mechanisms have been cited for histone deacetylase inhibitors: induction of gene expression, acetylation of cytoplasmic proteins and altered function, increased degradation of cytoplasmic proteins due to impaired Hsp90 activity, altered angiogenesis, and mitotic effects. Exactly which mechanism is of critical importance will be the subject of continuing investigation. Studies carried out in collaboration with the NCI drug screen suggest that a DNA damage fingerprint can be observed following romidepsin treatment. These studies have also been complemented by experiments aimed at identifying synergistic drug combinations. We have identified several two-drug combinations that have markedly increased the activity already observed in monotherapy in lymphoma and appear to translate into activity in solid tumors. Animal studies supporting a protocol concept for the combination of romidepsin with an experimental agent have been completed; we have submitted an LOI to study this combination in the clinic in the coming year. Tumor samples obtained from the xenografts in mice have demonstrated identical results to those obtained in the laboratory, in terms of response of pharmacodynamic markers to romidepsin exposure. In related studies, we have been impressed with the prevalent reduction in c-myc expression observed following romidepsin treatment. Given that myc is understood as a master regulator of cell metabolism, we will also undertake studies aimed at understanding the role of the myc response in the cell death that occurs following romidepsin. Finally, we have begun to evaluate combinations of epigenetic agents with the aim of developing a pharmacodynamic markers to use in clinical trials combining HDAC inhibitors and DNA methylation inhibitors. We now have multiple approaches for HDAC inhibitor combinations identified with excellent mechanistic data that we want to pursue in the clinic, first in Phase I and then in Phase II.